• DocumentCode
    2264886
  • Title

    Policy-directed data movement in grids

  • Author

    Feng, Jun ; Cui, Lingling ; Wasson, Glenn ; Humphrey, Marty

  • Author_Institution
    Dept. of Comput. Sci., Virginia Univ., Charlottesville, VA
  • Volume
    1
  • fYear
    0
  • fDate
    0-0 0
  • Abstract
    One of the guiding principles of the grid is local (site) autonomy. Resource owners maintain control over their resources even when those resources are part of a larger grid. In other words, resource sharing in grids must be subject to the policies of the (local) resource owners. To date, not enough attention has been paid to describing, manipulating or enforcing explicit resource usage policies. Most existing grid systems have either implicit resource usage policies (with ad-hoc enforcement mechanisms) or support only limited types of policies (e.g., security policies). Systems that do provide some support for resource usage policies typically consider only CPU resources, leaving the provisioning of other resources on the grid unconstrained. This paper focuses on policies for grid storage resources. We have identified classes of policies for grid storage resource providers and have implemented an explicit policy-based architecture to manage and enforce them. This architecture consists of two components, MyPolMan, a service to express and manage policies, and .NET GridFTP, an enforcement mechanism for policy-based grid data movement. We show how this system allows grid users to access storage resources through a familiar API while allowing local system administrators to control resource utilization
  • Keywords
    application program interfaces; grid computing; network operating systems; resource allocation; storage allocation; .NET GridFTP; API; CPU resources; MyPolMan; access storage resources; ad-hoc enforcement mechanisms; grid data movement; grid storage resources; grid systems; policy-directed data movement; resource sharing; resource usage policy; resource utilization; system administrators; Authentication; Collaborative work; Computer science; Control systems; Engineering profession; Grid computing; Large-scale systems; Resource management; Scheduling; Security;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Parallel and Distributed Systems, 2006. ICPADS 2006. 12th International Conference on
  • Conference_Location
    Minneapolis, MN
  • ISSN
    1521-9097
  • Print_ISBN
    0-7695-2612-8
  • Type

    conf

  • DOI
    10.1109/ICPADS.2006.82
  • Filename
    1655677